| spelling |
Adriano César Machado Pereirahttp://lattes.cnpq.br/6813736989856243Flávio Luis Cardeal PáduaHeitor Soares Ramos Filhohttp://lattes.cnpq.br/3188926301793732Johnathas Araújo de Carvalho2023-06-07T14:57:03Z2023-06-07T14:57:03Z2022-08-17http://hdl.handle.net/1843/54641Diversos trabalhos têm utilizado algoritmos de aprendizado de máquina e deep learning para realizar a predição de séries temporais financeiras ou prever a direção de movimento de preço de ativos, e demonstrado melhor desempenho em relação aos tradicionais métodos econométricos. Entretanto, ainda não há trabalhos que realizem um estudo sistemático do resultado histórico das predições a fim de estabelecer um sistema de gestão de risco (SGR) para operações automatizadas (algotrading). Este trabalho aplicou algoritmos de aprendizado profundo, especificamente as redes LSTM e Bi-LSTM, para realizar a predição de limites diários de negociação de um ativo (valores máximo e mínimo do ativo). Um estudo estatístico e sistemático das predições do modelo nos dados de treino permitiu estimar a probabilidade de acerto, relação de risco/ganho, bem como mensurar o grau de exposição adequado. As predições de limites foram utilizadas para filtrar as sugestões de operações de sete estratégias automatizadas baseadas em cruzamento de médias e osciladores (RSI e MACD). Simulações foram realizadas em 5 ativos mais negociados da B3 entre 2008 e 2019 (ABEV3, B3SA3, ITUB4, PETR4 e USIM5) e em 7 estratégias automatizadas. Os resultados indicaram que, apesar do grau de incerteza das predições dos modelos, a utilização do SGR proposto apresentou melhor desempenho em relação ao retorno líquido financeiro das estratégias em 64% das situações e ganho médio no retorno líquido de 0,31%. As estratégias que utilizaram SGR e que apresentaram melhor desempenho tiveram um retorno sobre o capital alocado superior ao baseline estabelecido de buy-and-hold.Machine learning and deep learning algorithms are being used by several studies to predict future values in financial series or the direction of assets price movements, and proved to outperform the traditional econometric models. Nevertheless, there are not yet working papers that conduct systematical analysis of historical predictions in order to establish a risk management system (RMS) for automated trades (algotrading). This work employ deep learning algorithms, specifically the LSTM and Bi-LSTM networks, to predict daily trading ranges of assets (high and low prices). A statistical and systematical analysis of the predictions allow to estimate the hit probability, risk/gain ratio, as well to measure more adequate exposure sizes. The range predictions were applied to filter suggested operations of seven automated trading strategies based on moving average and oscillatores (RSI and MACD). Backtestings were performed upon 5 most negotiated assets in B3 between 2008 and 2019 (ABEV3, B3SA3, ITUB4, PETR4 and USIM5). The results pointed that, despite of the predictions uncertainty of the models, the appliance of RMS outperformed the strategies in terms of net return in 64% of the cases and demonstrated average gain of 0,31% in the net return of the strategies. Those strategies that made use of risk management system and had the best performance for each tested asset also had a net return over allocated capital greater than the baseline defined as buy-and-hold strategy.porUniversidade Federal de Minas GeraisPrograma de Pós-Graduação em Ciência da ComputaçãoUFMGBrasilICX - DEPARTAMENTO DE CIÊNCIA DA COMPUTAÇÃOhttp://creativecommons.org/licenses/by-nc/3.0/pt/info:eu-repo/semantics/openAccessComputação – TesesAprendizado do computador – TesesGestão de riscos – TesesAlgoritmos de computador – TesesGestão de riscoAlgotradingAprendizado profundoSistema de gestão de risco em operações automatizadas na bolsa de valores baseado em aprendizado profundoRisk management system on automated trading for stock market based on deep learninginfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisreponame:Repositório Institucional da UFMGinstname:Universidade Federal de Minas Gerais (UFMG)instacron:UFMGLICENSElicense.txtlicense.txttext/plain; charset=utf-82118https://repositorio.ufmg.br/bitstream/1843/54641/5/license.txtcda590c95a0b51b4d15f60c9642ca272MD55ORIGINALDissertacao_Mestrado_Johnathas_Carvalho_vFINAL2.pdfDissertacao_Mestrado_Johnathas_Carvalho_vFINAL2.pdfapplication/pdf11412584https://repositorio.ufmg.br/bitstream/1843/54641/4/Dissertacao_Mestrado_Johnathas_Carvalho_vFINAL2.pdf4da9270328443623f770250f4dfbefd4MD54CC-LICENSElicense_rdflicense_rdfapplication/rdf+xml; charset=utf-8920https://repositorio.ufmg.br/bitstream/1843/54641/2/license_rdf33b8016dc5c4681c1e7a582a4161162cMD521843/546412023-06-07 11:57:04.363oai:repositorio.ufmg.br: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ório InstitucionalPUBhttps://repositorio.ufmg.br/oaiopendoar:2023-06-07T14:57:04Repositório Institucional da UFMG - Universidade Federal de Minas Gerais (UFMG)false
|